bench(coverage): image-mode so Word files are measurable (docx = 100% coverage)#141
Merged
Conversation
…measurable The coverage metric reported N/A for documents without figure captions. Add a second ground-truth mode: - image mode: for docs with < 2 captions, ground truth = embedded images in the source (office media / PDF xobjects, icons filtered). Covered = distinct images figmark extracted, deduplicated by content hash so the LibreOffice "same header on every page" phantom repeats (41 placements -> 6 unique) collapse to one. - Also hardened the caption regex to match any whitespace (incl. nbsp) between the word and number, so the ground truth is stable across publishers. Baseline now includes govuk-social-care-consultation.docx: 6/6 images = 100% coverage (the docx's issues are quality, not coverage). Caption docs unchanged (BoC 58%, BoJ 79%). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The coverage metric reported N/A for uncaptioned documents. This adds a second ground-truth mode so the Word file is in the benchmark.
media/or PDF xobjects, icons filtered by size). Covered = distinct images figmark extracted, deduped by content hash so LibreOffice's phantom repeats (41 placements → 6 unique) collapse to one.Baseline now
The docx has complete coverage — figmark finds all 6 embedded images; its open questions are quality (two logos over-described), which is the LLM-judge track. Scripts-only.
🤖 Generated with Claude Code